Preperation¶

Importing libraries & data & checking missing info¶

In [2]:
# your code
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline
In [3]:
#data for population from 2019 to 2020
df = pd.read_csv("HousingPrices-Amsterdam-August-2021.csv",sep = ',',header=0)
In [4]:
df=df.drop(columns=['Unnamed: 0'])
In [5]:
#no missing values
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 924 entries, 0 to 923
Data columns (total 7 columns):
 #   Column   Non-Null Count  Dtype  
---  ------   --------------  -----  
 0   Address  924 non-null    object 
 1   Zip      924 non-null    object 
 2   Price    920 non-null    float64
 3   Area     924 non-null    int64  
 4   Room     924 non-null    int64  
 5   Lon      924 non-null    float64
 6   Lat      924 non-null    float64
dtypes: float64(3), int64(2), object(2)
memory usage: 50.7+ KB
In [6]:
#describe basic info
df.describe()
Out[6]:
Price Area Room Lon Lat
count 9.200000e+02 924.000000 924.000000 924.000000 924.000000
mean 6.220654e+05 95.952381 3.571429 4.888605 52.363326
std 5.389942e+05 57.447436 1.592332 0.053140 0.024028
min 1.750000e+05 21.000000 1.000000 4.644819 52.291519
25% 3.500000e+05 60.750000 3.000000 4.855834 52.352077
50% 4.670000e+05 83.000000 3.000000 4.886818 52.364631
75% 7.000000e+05 113.000000 4.000000 4.922337 52.377598
max 5.950000e+06 623.000000 14.000000 5.029122 52.423805

Understanding the variables¶

Zip code (how many different neighborhoods?)¶

In [7]:
#Checking how many unique zip codes 
df["Zip"].nunique()
Out[7]:
834
In [11]:
# Getting evidence about the neighborhood, region, and street from zipcode
# make string version of original column, call it 'Zip1'
df['Zip1'] = df['Zip'].astype(str)

# make the new columns using string indexing
df['Region'] = df['Zip1'].str[0:2]
df['Neighbour'] = df['Zip1'].str[2:4]
df['Street'] = df['Zip1'].str[5:7]

# get rid of the extra variable (if you want)
df.drop('Zip1', axis=1, inplace=True)
In [12]:
df
Out[12]:
Address Zip Price Area Room Lon Lat Region Neighbour Street
0 Blasiusstraat 8 2, Amsterdam 1091 CR 685000.0 64 3 4.907736 52.356157 10 91 CR
1 Kromme Leimuidenstraat 13 H, Amsterdam 1059 EL 475000.0 60 3 4.850476 52.348586 10 59 EL
2 Zaaiersweg 11 A, Amsterdam 1097 SM 850000.0 109 4 4.944774 52.343782 10 97 SM
3 Tenerifestraat 40, Amsterdam 1060 TH 580000.0 128 6 4.789928 52.343712 10 60 TH
4 Winterjanpad 21, Amsterdam 1036 KN 720000.0 138 5 4.902503 52.410538 10 36 KN
... ... ... ... ... ... ... ... ... ... ...
919 Ringdijk, Amsterdam 1097 AE 750000.0 117 1 4.927757 52.354173 10 97 AE
920 Kleine Beerstraat 31, Amsterdam 1033 CP 350000.0 72 3 4.890612 52.414587 10 33 CP
921 Stuyvesantstraat 33 II, Amsterdam 1058 AK 350000.0 51 3 4.856935 52.363256 10 58 AK
922 John Blankensteinstraat 51, Amsterdam 1095 MB 599000.0 113 4 4.965731 52.375268 10 95 MB
923 S. F. van Ossstraat 334, Amsterdam 1068 JS 300000.0 79 4 4.810678 52.355493 10 68 JS

924 rows × 10 columns

Unique Neighborhoods¶

In [14]:
df["Neighbour"].unique()
Out[14]:
array(['91', '59', '97', '60', '36', '51', '33', '53', '58', '73', '19',
       '17', '64', '02', '13', '16', '72', '62', '95', '34', '35', '83',
       '56', '65', '78', '76', '54', '18', '06', '93', '04', '24', '68',
       '75', '11', '55', '94', '07', '67', '25', '52', '57', '15', '61',
       '12', '77', '98', '63', '92', '03', '21', '69', '96', '66', '82',
       '87', '81', '27', '71', '08', '32', '28', '86', '14', '31', '23',
       '74', '22', '79', '09'], dtype=object)

Top most frequent neighborhoods¶

In [10]:
df.Neighbor.value_counts().head(10).sort_values(ascending=False)
Out[10]:
17    32
55    29
68    26
53    26
18    25
58    25
02    25
13    24
19    24
57    23
Name: Neighbor, dtype: int64

Part 1) Neighborhoods of the highest housing price (top 10)¶

Q1. Where is the most expensive home? And what's the price?¶

In [ ]:
# Sorting by column 'Price'
price=df.sort_values(by=['Price'], ascending=False)
In [16]:
price=price.head(10)
price
Out[16]:
Address Zip Price Area Room Lon Lat Region Neighbour Street
195 Keizersgracht 544, Amsterdam 1017 EL 5950000.0 394 10 4.887378 52.365078 10 17 EL
837 Oranje Nassaulaan 9, Amsterdam 1075 AH 5850000.0 480 14 4.863729 52.354874 10 75 AH
305 Singel 318, Amsterdam 1016 AE 4900000.0 623 13 4.887869 52.371039 10 16 AE
103 Westeinde 16, Amsterdam 1017 ZP 4550000.0 497 13 4.898620 52.358798 10 17 ZP
179 Dam 5 V, Amsterdam 1012 JS 4495000.0 178 5 4.894290 52.373106 10 12 JS
33 Vondelstraat 51 hs, Amsterdam 1054 GJ 3925000.0 319 7 4.875471 52.361571 10 54 GJ
301 Badhuiskade 353, Amsterdam 1031 KV 3680000.0 374 4 4.905695 52.385321 10 31 KV
276 Sophialaan 45, Amsterdam 1075 BM 3500000.0 374 7 4.857608 52.352366 10 75 BM
334 Singel 146 A, Amsterdam 1015 AG 3500000.0 348 8 4.889816 52.375532 10 15 AG
264 Olympiakade 12, Amsterdam 1076 XP 2750000.0 309 8 4.862413 52.349528 10 76 XP
  • The most expensive house, 5,950,000 euros is located in the postal code area 1017, which is near to the centre and DePijp
  • The second most expensive house, 5,850,000 is located in the postal code area 1075, right down the Vondelpark.
In [14]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[14]:

Part 1-b) Top 10 most expensive neighborhoods further analysis¶

Q2. Where are the top 10 highest houses located in terms of neighborhoods?¶

In [19]:
Neighbordict=dict(price['Neighbour'].value_counts())
Neighbordict
Out[19]:
{'17': 2, '75': 2, '16': 1, '12': 1, '54': 1, '31': 1, '15': 1, '76': 1}
In [20]:
#The top 10 highest housing price Neighborhoods (in a dataframe)
pd.DataFrame(price['Neighbour'].value_counts().nlargest(10))
Out[20]:
Neighbour
17 2
75 2
16 1
12 1
54 1
31 1
15 1
76 1
  • The top 10 expensive houses were located in 8 different neighborhoods
  • They are located in those neighbourhoods: 1017, 1075, 1016, 1012, 1054, 1031, 1015, 1076
In [24]:
#sorting houses in the top 10 housing prices neighborhoods
#creating dataframe only with those neighborcodes above (neighborhood of the top 10 housing price )
Nb_sorted=df[df["Neighbour"].isin(Neighbordict)]
Nb_sorted
Out[24]:
Address Zip Price Area Room Lon Lat Region Neighbour Street
13 Amstel 124 B, Amsterdam 1017 AD 575000.0 70 2 4.897782 52.366928 10 17 AD
20 Keizersgracht 171 F, Amsterdam 1016 DP 1625000.0 199 6 4.885830 52.374742 10 16 DP
32 Van Tuyll van Serooskerkenweg 86 H, Amsterdam 1076 JP 735000.0 83 3 4.861090 52.344643 10 76 JP
33 Vondelstraat 51 hs, Amsterdam 1054 GJ 3925000.0 319 7 4.875471 52.361571 10 54 GJ
37 J.J. Cremerplein 24 II, Amsterdam 1054 TL 570000.0 63 3 4.858560 52.359929 10 54 TL
... ... ... ... ... ... ... ... ... ... ...
902 Lauriergracht 89, Amsterdam 1016 RJ 1300000.0 139 5 4.879252 52.370823 10 16 RJ
906 Buiksloterweg 99, Amsterdam 1031 CJ 1250000.0 180 9 4.914882 52.388128 10 31 CJ
910 Valeriusstraat 193, Amsterdam 1075 EW 1698000.0 205 5 4.860637 52.351894 10 75 EW
914 Hagedoornplein 24, Amsterdam 1031 BV 849000.0 111 5 4.913690 52.388160 10 31 BV
916 Eerste Looiersdwarsstraat 10 B, Amsterdam 1016 VM 525000.0 59 4 4.880977 52.368961 10 16 VM

137 rows × 10 columns

Q3. What is the average price of those most expensive neighborhoods?¶

Top 10 highest housing price neighbourhoods average housing price, areas & the number of rooms in Amsterdam¶

In [29]:
#sorting out the most expensive housing neighborhood grouped by average prices
top10_avg_price= df.groupby('Neighbour').mean().sort_values(by=['Price'],ascending = False).reset_index().head(10)
top10_avg_price
Out[29]:
Neighbour Price Area Room Lon Lat
0 75 1.509654e+06 148.769231 5.000000 4.859277 52.352815
1 31 1.395600e+06 171.200000 5.000000 4.908616 52.387439
2 17 1.259000e+06 131.187500 4.000000 4.893137 52.363220
3 71 1.256500e+06 123.000000 3.600000 4.880682 52.356953
4 16 1.168632e+06 140.894737 4.578947 4.869861 52.371341
5 28 1.097500e+06 158.500000 4.500000 5.011931 52.402957
6 12 1.093533e+06 113.666667 3.533333 4.894240 52.372861
7 27 1.067000e+06 217.666667 5.333333 4.965498 52.410423
8 77 9.820000e+05 136.894737 4.473684 4.873186 52.346582
9 14 9.630000e+05 108.800000 3.000000 4.881205 52.393870
  • The most expensive neighbourhood were (in order) 1075 (Willemspark), 1031,1017,1071, 1016, 1028, 1012, 1027, 1077, and 1014.
  • The average price of the most expensive neighbourhood (1075-(Willemspark)) is 1,509,654 euros.
In [26]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[26]:
In [31]:
sns.set(rc={'figure.figsize':(11.7,8.27)})
sns.barplot(data = top10_avg_price, x='Neighbour', y='Price').set(title='Most expensive average housing neighborhoods')
Out[31]:
[Text(0.5, 1.0, 'Most expensive average housing neighborhoods')]

Q4. Is the most expensive house located in the highest average housing neighborhoods?¶

The top 10 expensive houses in Amsterdam¶

In [30]:
#sorting out the most expensive housing neighborhood grouping by average prices
top10_price= df.groupby('Neighbour').max().sort_values(by=['Price'],ascending = False).reset_index().head(10)
top10_price
Out[30]:
Neighbour Address Zip Price Area Room Lon Lat Region Street
0 17 Weteringschans 122 A, Amsterdam 1017 ZP 5950000.0 497 13 4.901713 52.368029 10 ZP
1 75 Valeriusstraat 193, Amsterdam 1075 XR 5850000.0 480 14 4.865847 52.354874 10 XR
2 16 Singel 318, Amsterdam 1016 XA 4900000.0 623 13 4.888199 52.378060 10 XA
3 12 Wijdesteeg 62, Amsterdam 1012 XJ 4495000.0 213 7 4.901320 52.376702 10 XJ
4 54 Zocherstraat 23 X, Amsterdam 1054 ZT 3925000.0 319 7 4.880571 52.365121 10 ZT
5 31 Spadinalaan 53, Amsterdam 1031 VD 3680000.0 374 9 4.914882 52.388684 10 VD
6 15 Westerstraat 268, Amsterdam 1015 ZH 3500000.0 348 8 4.893043 52.382746 10 ZH
7 76 Van Tuyll van Serooskerkenweg 86 H, Amsterdam 1076 XP 2750000.0 309 8 4.864958 52.349528 10 XP
8 73 Van Ostadestraat 193 H, Amsterdam 1073 VP 2650000.0 387 10 4.902864 52.355854 10 VP
9 71 Willemsparkweg 125 C, Amsterdam 1071 VE 2500000.0 202 6 4.885645 52.361226 10 VE
  • It doesn't necessarily seem the case that expensive houses are located in the highest average housing neighbourhood.

Part 2) Residents info¶

Who are living in those area? And where are those places?¶

Preperation¶

Source¶

In [24]:
#Found this dataset from the dutch official open data (https://opendata.cbs.nl/statline/portal.html?_la=nl&_catalog=CBS&tableId=83504NED&_theme=270)
#("https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83504NED/table?dl=76864")

Importing data & pre-processing¶

In [32]:
info=pd.read_excel('top10_status.xlsx', header=0)
In [33]:
info
Out[33]:
Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 1012 4105 185 485 2900
1 1014 510 55 770 750
2 1016 4630 290 1050 3430
3 1017 4790 295 1570 4005
4 1027 105 15 135 125
5 1028 75 20 190 125
6 1031 2055 420 1135 1515
7 1071 3495 380 2320 3395
8 1075 3120 415 1835 2735
9 1077 1890 285 2545 2080

Spliting the zipcode again

In [39]:
# Getting evidence about the neighborhood, region, and street from zipcode
# make string version of original column, call it 'Zip1'
info['Zip1'] = info['Postal code'].astype(str)

# make the new columns using string indexing
info['Street'] = info['Zip1'].str[0:2]
info['Neighbour'] = info['Zip1'].str[2:4]

# get rid of the extra variable (if you want)
info.drop('Zip1', axis=1, inplace=True)
In [40]:
info
Out[40]:
Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren) Street Neighbor Neighbour
0 1012 4105 185 485 2900 10 12 12
1 1014 510 55 770 750 10 14 14
2 1016 4630 290 1050 3430 10 16 16
3 1017 4790 295 1570 4005 10 17 17
4 1027 105 15 135 125 10 27 27
5 1028 75 20 190 125 10 28 28
6 1031 2055 420 1135 1515 10 31 31
7 1071 3495 380 2320 3395 10 71 71
8 1075 3120 415 1835 2735 10 75 75
9 1077 1890 285 2545 2080 10 77 77

Merge

In [41]:
# combining the original table with the top 10 most expensive neighborhoods info table
df_merged = pd.merge(df, info, right_on='Neighbour', left_on='Neighbour', how='inner').drop(columns=['Lon', 'Lat'])
df_merged
Out[41]:
Address Zip Price Area Room Region Neighbour Street_x Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren) Street_y Neighbor
0 Amstel 124 B, Amsterdam 1017 AD 575000.0 70 2 10 17 AD 1017 4790 295 1570 4005 10 17
1 Vijzelstraat 91, Amsterdam 1017 HH 1185000.0 160 5 10 17 HH 1017 4790 295 1570 4005 10 17
2 Falckstraat 47 A, Amsterdam 1017 VV NaN 147 3 10 17 VV 1017 4790 295 1570 4005 10 17
3 Maarten Jansz. Kosterstraat 10, Amsterdam 1017 VX 650000.0 67 4 10 17 VX 1017 4790 295 1570 4005 10 17
4 Westeinde 16, Amsterdam 1017 ZP 4550000.0 497 13 10 17 ZP 1017 4790 295 1570 4005 10 17
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
121 Badhuiskade 353, Amsterdam 1031 KV 3680000.0 374 4 10 31 KV 1031 2055 420 1135 1515 10 31
122 Spadinalaan 53, Amsterdam 1031 KB 650000.0 90 3 10 31 KB 1031 2055 420 1135 1515 10 31
123 Magnoliastraat 10, Amsterdam 1031 VD 549000.0 101 4 10 31 VD 1031 2055 420 1135 1515 10 31
124 Buiksloterweg 99, Amsterdam 1031 CJ 1250000.0 180 9 10 31 CJ 1031 2055 420 1135 1515 10 31
125 Hagedoornplein 24, Amsterdam 1031 BV 849000.0 111 5 10 31 BV 1031 2055 420 1135 1515 10 31

126 rows × 15 columns

In [42]:
#Checking whether the 10 unique zip codes are in the data frame
df_merged.Neighbor.nunique()
Out[42]:
10

Q5. What are some characteristics of occupants living in the top 10 most expensive neighborhoods?¶

In [43]:
#same as above, sorting out price
#sorting out the most expensive housing neighborhood grouping by average prices
top10_neigh_status= df_merged.groupby('Neighbour').mean().sort_values(by=['Price'],ascending = False).reset_index().head(10)
top10_neigh_status
Out[43]:
Neighbour Price Area Room Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 75 1.509654e+06 148.769231 5.000000 1075.0 3120.0 415.0 1835.0 2735.0
1 31 1.395600e+06 171.200000 5.000000 1031.0 2055.0 420.0 1135.0 1515.0
2 17 1.259000e+06 131.187500 4.000000 1017.0 4790.0 295.0 1570.0 4005.0
3 71 1.256500e+06 123.000000 3.600000 1071.0 3495.0 380.0 2320.0 3395.0
4 16 1.168632e+06 140.894737 4.578947 1016.0 4630.0 290.0 1050.0 3430.0
5 28 1.097500e+06 158.500000 4.500000 1028.0 75.0 20.0 190.0 125.0
6 12 1.093533e+06 113.666667 3.533333 1012.0 4105.0 185.0 485.0 2900.0
7 27 1.067000e+06 217.666667 5.333333 1027.0 105.0 15.0 135.0 125.0
8 77 9.820000e+05 136.894737 4.473684 1077.0 1890.0 285.0 2545.0 2080.0
9 14 9.630000e+05 108.800000 3.000000 1014.0 510.0 55.0 770.0 750.0

  • The most expensive neighbourhood is located in 1075 (Willemspark) postal code area with the average housing price of 1,509,654 euros: The biggest proportion was single, followed by partner with no children, partner with children and single parents. Bettwen partners and single parents, there's a huge gap.

  • The second most expensive neighbourhood is in the postal code 1031 with the average housing price of 1,395,600 euros. This area has been developed recently with lots of new apartments. Also this place is popular for single and no children couples.

  • The third most expensive neighborhood is located in the postal code 1017 (Grachtengordel), where it's near to the city centre and DePijip (very hip & old district): The biggest proportion is single, followed by partner with no children, partner with children and parents in sigle parents. Between partners and single parents, there's a huge gap.

  • The fourth most expensive neighborhood is 1071, where is the neighbourhood district to 1075 (Willemspark).

  • The fifth most expensive neighborhood is 1016, where the neighbourhood distrcits are central, Jordann and Vondelpark.

  • Among the 10 expensive neighbourhood, for people with kid(s), the neighborhood 1077 (Amsterdam-Zuid) is most popular, whereas for people with no kids (single & partner with no children), the most popular areas are in the city centre.

In [33]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[33]:

Part 3) Where do you want to live?¶

Based on the insights above, some people might be interested in living in one of those top 10 neighbourhoods. But depending on your current status (single/couple/single parent), you might want to know which of those areas are popular among those different types of housing occupant.

Q6. Where do single people, partners with and without children live the most in those most expensive housing area in the city?¶

Single occupants¶

In [34]:
#sorting out the most expensive housing neighborhood grouping by average prices
top10_neigh_status_single= top10_neigh_status.groupby('Neighbor').sum().sort_values(by=['Single'],ascending = False).reset_index().head(10)
top10_neigh_status_single
Out[34]:
Neighbor Price Area Room Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 17 1.259000e+06 131.187500 4.000000 1017.0 4790.0 295.0 1570.0 4005.0
1 16 1.168632e+06 140.894737 4.578947 1016.0 4630.0 290.0 1050.0 3430.0
2 12 1.093533e+06 113.666667 3.533333 1012.0 4105.0 185.0 485.0 2900.0
3 71 1.256500e+06 123.000000 3.600000 1071.0 3495.0 380.0 2320.0 3395.0
4 75 1.509654e+06 148.769231 5.000000 1075.0 3120.0 415.0 1835.0 2735.0
5 31 1.395600e+06 171.200000 5.000000 1031.0 2055.0 420.0 1135.0 1515.0
6 77 9.820000e+05 136.894737 4.473684 1077.0 1890.0 285.0 2545.0 2080.0
7 14 9.630000e+05 108.800000 3.000000 1014.0 510.0 55.0 770.0 750.0
8 27 1.067000e+06 217.666667 5.333333 1027.0 105.0 15.0 135.0 125.0
9 28 1.097500e+06 158.500000 4.500000 1028.0 75.0 20.0 190.0 125.0
In [35]:
sns.set(rc={'figure.figsize':(11.7,8.27)})
sns.barplot(data = top10_neigh_status_single, x='Neighbor', y='Single')
Out[35]:
<AxesSubplot:xlabel='Neighbor', ylabel='Single'>
In [36]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[36]:

Among the top 10 expensive neighbourhoods, single occupants live the most in the postal code area 1017, followed by 1016 and 1012. Those are the areas cover Amsterdam-Centrum, Centrum, Grachtengordel-Zuid, Stadsdeel Centrum, Centrum, De Wallen.

Couple with child(ren)¶

In [46]:
#sorting out the most expensive housing neighborhood grouping by average prices
top10_neigh_status_wchild= top10_neigh_status.groupby('Neighbour').sum().sort_values(by=['Partner in couple with child(ren)'],ascending = False).reset_index().head(10)
top10_neigh_status_wchild
Out[46]:
Neighbour Price Area Room Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 77 9.820000e+05 136.894737 4.473684 1077.0 1890.0 285.0 2545.0 2080.0
1 71 1.256500e+06 123.000000 3.600000 1071.0 3495.0 380.0 2320.0 3395.0
2 75 1.509654e+06 148.769231 5.000000 1075.0 3120.0 415.0 1835.0 2735.0
3 17 1.259000e+06 131.187500 4.000000 1017.0 4790.0 295.0 1570.0 4005.0
4 31 1.395600e+06 171.200000 5.000000 1031.0 2055.0 420.0 1135.0 1515.0
5 16 1.168632e+06 140.894737 4.578947 1016.0 4630.0 290.0 1050.0 3430.0
6 14 9.630000e+05 108.800000 3.000000 1014.0 510.0 55.0 770.0 750.0
7 12 1.093533e+06 113.666667 3.533333 1012.0 4105.0 185.0 485.0 2900.0
8 28 1.097500e+06 158.500000 4.500000 1028.0 75.0 20.0 190.0 125.0
9 27 1.067000e+06 217.666667 5.333333 1027.0 105.0 15.0 135.0 125.0
In [47]:
sns.set(rc={'figure.figsize':(11.7,8.27)})
sns.barplot(data = top10_neigh_status_wchild, x='Neighbour', y='Partner in couple with child(ren)')
Out[47]:
<AxesSubplot:xlabel='Neighbour', ylabel='Partner in couple with child(ren)'>
In [39]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[39]:

Among the top 10 expensive neighbourhoods, couples with children live the most in the postal code area 1077, followed by 1071 and 1075. Those are the areas are closed to each other near to the Vondelpark, covering Ood-Zuid,Amsterdam-Zuid, and Museumkwartier.

No children couple¶

In [48]:
#sorting out the most expensive housing neighborhood grouping by average prices
top10_neigh_status_wnochild= top10_neigh_status.groupby('Neighbour').sum().sort_values(by=['Partner in couple, no child(ren)'],ascending = False).reset_index().head(10)
top10_neigh_status_wnochild
Out[48]:
Neighbour Price Area Room Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 17 1.259000e+06 131.187500 4.000000 1017.0 4790.0 295.0 1570.0 4005.0
1 16 1.168632e+06 140.894737 4.578947 1016.0 4630.0 290.0 1050.0 3430.0
2 71 1.256500e+06 123.000000 3.600000 1071.0 3495.0 380.0 2320.0 3395.0
3 12 1.093533e+06 113.666667 3.533333 1012.0 4105.0 185.0 485.0 2900.0
4 75 1.509654e+06 148.769231 5.000000 1075.0 3120.0 415.0 1835.0 2735.0
5 77 9.820000e+05 136.894737 4.473684 1077.0 1890.0 285.0 2545.0 2080.0
6 31 1.395600e+06 171.200000 5.000000 1031.0 2055.0 420.0 1135.0 1515.0
7 14 9.630000e+05 108.800000 3.000000 1014.0 510.0 55.0 770.0 750.0
8 27 1.067000e+06 217.666667 5.333333 1027.0 105.0 15.0 135.0 125.0
9 28 1.097500e+06 158.500000 4.500000 1028.0 75.0 20.0 190.0 125.0
In [49]:
sns.set(rc={'figure.figsize':(11.7,8.27)})
sns.barplot(data = top10_neigh_status_wnochild, x='Neighbour', y='Partner in couple, no child(ren)')
Out[49]:
<AxesSubplot:xlabel='Neighbour', ylabel='Partner in couple, no child(ren)'>
In [42]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[42]:

Among the top 10 expensive neighbourhoods, couples with no children live the most in the postal code area 1017, followed by 1016 and 1071. This result is very akin to single occupants, indicating that in those three areas there are many single and couple with no children. Those are the areas cover Amsterdam-Centrum, Centrum, Grachtengordel-Zuid, Stadsdeel Centrum, Museumkwartier, and Oud-Zuid.

Single parents¶

In [50]:
#sorting out the most expensive housing neighborhood grouping by single parents
top10_neigh_status_singleparents= top10_neigh_status.groupby('Neighbour').sum().sort_values(by=['Parent in a single-parent household'],ascending = False).reset_index().head(10)
top10_neigh_status_singleparents
Out[50]:
Neighbour Price Area Room Postal code Single Parent in a single-parent household Partner in couple with child(ren) Partner in couple, no child(ren)
0 31 1.395600e+06 171.200000 5.000000 1031.0 2055.0 420.0 1135.0 1515.0
1 75 1.509654e+06 148.769231 5.000000 1075.0 3120.0 415.0 1835.0 2735.0
2 71 1.256500e+06 123.000000 3.600000 1071.0 3495.0 380.0 2320.0 3395.0
3 17 1.259000e+06 131.187500 4.000000 1017.0 4790.0 295.0 1570.0 4005.0
4 16 1.168632e+06 140.894737 4.578947 1016.0 4630.0 290.0 1050.0 3430.0
5 77 9.820000e+05 136.894737 4.473684 1077.0 1890.0 285.0 2545.0 2080.0
6 12 1.093533e+06 113.666667 3.533333 1012.0 4105.0 185.0 485.0 2900.0
7 14 9.630000e+05 108.800000 3.000000 1014.0 510.0 55.0 770.0 750.0
8 28 1.097500e+06 158.500000 4.500000 1028.0 75.0 20.0 190.0 125.0
9 27 1.067000e+06 217.666667 5.333333 1027.0 105.0 15.0 135.0 125.0
In [44]:
sns.set(rc={'figure.figsize':(11.7,8.27)})
sns.barplot(data = top10_neigh_status_singleparents, x='Neighbor', y='Parent in a single-parent household')
Out[44]:
<AxesSubplot:xlabel='Neighbor', ylabel='Parent in a single-parent household'>
In [45]:
from IPython import display
display.Image("https://slotenmaker365-amsterdam.nl/wp-content/uploads/2018/04/map-amsterdam.png",width=500, height=500)
Out[45]:

Among the top 10 expensive neighbourhoods, single parents live the most in the postal code area 1031, followed by 1075 and 1071. The most popular place 1031 is in the Noord. This result is different from the other three, which the most number of occupants were found around the city. Then the second and the third most popular places for single parents are 1075 nad 1071 where are also known as Zuid part of Amsterdam.

Questions from the analysis¶

  • Is the criminal rate lower in Zuid and Noord? Since it should be safe to raise the kids for single parents.
  • Average age of occupants living in 1017 and 1016 (where single & couple with no children live the most) compared to 1077 and and 1071 (couple with children live the most)

Addition¶

Visualising Amsterdam city in terms of housing price¶

In [50]:
# install plotly and shapely if not already installed

!pip install plotly
!pip install shapely

Requirement already satisfied: plotly in /Users/joyena/opt/anaconda3/lib/python3.9/site-packages (5.6.0)
Requirement already satisfied: six in /Users/joyena/opt/anaconda3/lib/python3.9/site-packages (from plotly) (1.16.0)
Requirement already satisfied: tenacity>=6.2.0 in /Users/joyena/opt/anaconda3/lib/python3.9/site-packages (from plotly) (8.0.1)
Collecting shapely
  Downloading shapely-2.0.0-cp39-cp39-macosx_10_9_x86_64.whl (1.4 MB)
     |████████████████████████████████| 1.4 MB 1.0 MB/s eta 0:00:01
Requirement already satisfied: numpy>=1.14 in /Users/joyena/opt/anaconda3/lib/python3.9/site-packages (from shapely) (1.21.5)
Installing collected packages: shapely
Successfully installed shapely-2.0.0
In [51]:
from urllib.request import urlopen
import json
with open('amsterdam.geojson') as response:
    counties = json.load(response)
counties['features'][0]
Out[51]:
{'type': 'Feature',
 'id': 0,
 'properties': {'name': 'Burgwallen-Oude Zijde'},
 'geometry': {'type': 'Polygon',
  'coordinates': [[[4.895492032180276, 52.367518292844046],
    [4.895246533269752, 52.367566817799876],
    [4.894707275231997, 52.36725908335511],
    [4.894092041776886, 52.367321279055304],
    [4.892775618450123, 52.36841912114592],
    [4.892450955103008, 52.36907274473733],
    [4.893051943943746, 52.37217664450764],
    [4.892905995889044, 52.37230060497486],
    [4.892951363414091, 52.37272294705677],
    [4.893152803521489, 52.37320031602717],
    [4.896530782872414, 52.37572206643713],
    [4.897022706563314, 52.37549968093607],
    [4.899348875669522, 52.377188009854805],
    [4.901636829193176, 52.37729232386225],
    [4.902728000440081, 52.37707681937856],
    [4.903157185509098, 52.376754792818375],
    [4.902888086171139, 52.37639679110526],
    [4.902452168376638, 52.37622644314114],
    [4.90123313917678, 52.37418038492089],
    [4.900917477297648, 52.37357848579406],
    [4.900737231610852, 52.373149166796516],
    [4.901039079201679, 52.373095042269966],
    [4.900433371387384, 52.3720618887445],
    [4.899863427984754, 52.37218658553432],
    [4.898713692355606, 52.37098261801167],
    [4.895492032180276, 52.367518292844046]]]}}
In [52]:
from shapely.geometry import shape, Point

def map_code(lon, lat):
    for feature in counties['features']:
        polygon = shape(feature['geometry'])
        p = Point(lon, lat)
        if polygon.contains(p):
            return feature['id']
In [53]:
df.columns
Out[53]:
Index(['Address', 'Zip', 'Price', 'Area', 'Room', 'Lon', 'Lat', 'Region',
       'Neighbor', 'Street'],
      dtype='object')
In [54]:
df['id'] =  df.apply(func=lambda row: map_code(row['Lon'], row['Lat']), axis=1)
In [55]:
df['Price'] = df['Price']/1e6
In [56]:
df_grouped = df.groupby('id').mean(numeric_only=True)
df_grouped.describe()
Out[56]:
Price Area Room Lon Lat
count 88.000000 88.000000 88.000000 88.000000 88.000000
mean 0.641881 95.858653 3.552620 4.892169 52.363383
std 0.345765 32.419766 0.786875 0.050536 0.024386
min 0.255000 45.800000 2.312500 4.783823 52.295265
25% 0.422438 72.864583 3.000000 4.861243 52.351645
50% 0.538312 91.203846 3.439614 4.890675 52.363012
75% 0.715571 110.896825 3.937500 4.922145 52.377223
max 2.179125 213.125000 6.625000 5.015949 52.420803
In [59]:
import plotly.express as px

fig = px.choropleth(df, geojson=counties, locations='id', color='Price',
                           color_continuous_scale="Viridis",
                           range_color=(0, 2),
                           scope="europe",
                           labels={'Price':'Price (millions €)'}
                          )

fig.update_geos(fitbounds="locations", visible=False)
fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()
In [35]:
from IPython import display
display.Image("map.png")
Out[35]:
In [ ]: